logo

Select Sidearea

Populate the sidearea with useful widgets. It’s simple to add images, categories, latest post, social media icon links, tag clouds, and more.

[email protected]
+1234567890

Probability Matters

It is well established that Americans do not understand probability.  There is example after example after example after example to demonstrate that fact. 


As admissions professionals, we must put serious effort into overcoming this challenge.  I would put it forward as a professional necessity – a core skill we must have mastered before rising to leadership.


Let me ask you a question: if your college has a yield rate of 17%, how many admitted students do you need for one enrolled student?


If you’re like me, you do the quick math.  17% is roughly one out of six, so I would need six admits for every enrolled student; that is my goal.


Of course, that’s absolutely the wrong way of thinking about it.  Even if the yield rate did equal probability (it does not) – each student would individually have a one in six chance of enrolling.  So if the first five students don’t enroll, what is the likelihood that the sixth one will?


(It’s still just one-in-six.)


I’ve talked about Nate Silver before – and feel oddly defensive of his election forecast model built on probability.  He has unfairly faced criticism that his probabilities were wrong since, for example, it said Donald Trump only had a one-in-six chance of winning the 2016 election.  That Donald Trump won was seen as proof that the model, showing an 84% chance that he would lose, was wrong.


That’s now how probability works, not in elections, not in poker, and not in admissions.


Think of it this way: when flipping a coin, what is the probability it lands heads up? 


One in two.


So if you flip it once and it lands tails, what is the probability it lands heads up the next time?


We all know it is still one in two.


Even if the coin landed tails seven times in a row, the next flip still has a one-in-two probability of landing heads up.


The point is – that a projected yield rate is not a per-student probability that you can build on.  It is an anticipated outcome, but each student needs attention, data analysis, and an assessment of their probability of enrollment.


(Also – the probability of the coin coming up tails eight times in a row is one-in-two-hundred-and-fifty-six.)

Teege Mettille
No Comments

Post a Comment

Comment
Name
Email
Website